knitr::opts_chunk$set(
    echo = TRUE,
    error = FALSE,
    message = FALSE,
    warning = FALSE
)
library(tidyverse)
library(kableExtra)
library(knitr)
library(plotly)
library(broom)
library(wesanderson)
library(robust)
library(ggparl)

# reads excel data sheet

data <- read.csv("../data/study4.csv") # reads data
data_sat <- read.csv("../data/raw/stage3_satisfaction.csv") # reads in satisfaction data



# summary(data) # explore the data types, missing data and typos

data <- data %>%
  mutate(gender = ifelse(gender == "m", "m", "f")) # corrects an issuewith extra space in "f "

data$gender <- as.factor(data$gender)
data$pp_number <- as.factor(data$pp_number)
data$condition <-as.factor(data$condition)


# mood data
mood1 <- read.csv("../data/raw/stage1_mood.csv") %>%
  select(subject, stimulusitem2, response)%>%
  rename(response1=response, feeling = stimulusitem2)%>%
  mutate(stage = 1)

mood2 <- read.csv("../data/raw/stage2_mood.csv") %>%
  select(subject, stimulusitem2,  response)%>%
  rename(response2=response,  feeling = stimulusitem2)%>%
  mutate(stage = 2)

mood3 <- read.csv("../data/raw/stage3_mood.csv") %>%
  select(subject, stimulusitem2, response) %>%
  rename(response3=response,  feeling = stimulusitem2)%>%
  mutate(stage = 3)


# qualtrics data 
      #mainly info about drinking

data_q <- read.csv("../data/other/study_4Q.csv")%>%
  select(Q21, gender, age, Q15, Q16)%>%
  rename(subject = Q21, 
         units = Q15, 
         units_beer = Q16)%>%
  mutate(subject = as.numeric(subject),
         condition = subject %/% 1000)%>%
  filter(!is.na(subject))%>%
  mutate(subject = as.factor(subject), 
         gender=as.factor(gender), 
         age = as.numeric(age), 
         units = as.numeric(units), 
         units_beer = as.numeric(units_beer))%>%
  filter( subject != "6666")

DA <- data %>%
  select(pp_number, BMI)%>%
  rename(subject = pp_number)

DB <- data_sat %>%
  select(subject, response, trialcode ) %>%
  mutate(row = row_number(), subject = as.factor(subject)) %>%
  pivot_wider(names_from = trialcode, values_from = response) %>%
  rename( wtp = VAS_wtp)%>%
  select(- row)


D1 <- inner_join( DA,DB,
                  by = "subject",
                  keep = FALSE
                ) %>%
    pivot_longer(cols =c(VAS, wtp) , 
                 names_to = "measure", 
                 values_to = "rating") %>%
    inner_join(., data_q,
             by = "subject", 
             keep = FALSE)%>%
  filter(!is.na(rating))
                                            # long dataset incl,. info about satisfaction and willingness to pay by condition, and alcohol/ non-alcohol
# need to add focus condition


#Joining mood data

D2 <- inner_join(mood1, mood2, 
                 by = c("subject", "feeling"),
                 keep = FALSE) %>%
  inner_join(., mood3,
             by = c("subject","feeling"),
             keep = FALSE)%>%
            select(- starts_with("stage")) %>%
  mutate(condition = subject  %/% 1000) %>%
  pivot_longer(cols = starts_with("response"),
   names_to = "stage",
   names_prefix = "response",
   values_to = "rating")

# gender
total_n <- nrow(data)
males <- data_q%>%filter(gender == "male") %>% nrow()
females <- data_q%>%filter(gender == "female") %>% nrow()
nonbinary <- data_q %>% filter(gender== "non-binary" )%>% nrow()



# by condition
gender_tib <- D1 %>% 
  filter(measure == "wtp") %>% # avoids duplicating data
  group_by( condition) %>%
  summarise(n=n(),
            BMI = mean(BMI, na.rm = TRUE), 
            age = mean(age, na.rm = TRUE)) %>% 
  kable()